Skip to content

Guides and resources for AI engineers: an agents course, five handbooks, a hands-on second-brain project and a vetted reading list.

1.6 KB
Raw Blame

plugins

Claude Code plugins that ship the course's tools. This repo is a plugin marketplace: add it once, install what you need.

claude plugin marketplace add undefined-ui/second-brain-os
claude plugin install agents-course@second-brain-os

agents-course

Each tool does the practice page of one course module, in your own repo.

Tool Invoke as Module
Context audit /agents-course:context-audit 1 · Context — measure the four places, find cache killers and window bloat
Goal test /agents-course:goal-test 2 · Loop — turn a task into a testable done, generate the script and the bounded loop
Gate check /agents-course:gate-check 3 · The gate — find the decisions that do not need the big model, propose fail-closed gates
Harness audit /agents-course:harness-audit 4 · Harness — walk the four rings, state the blast radius, rank the hardening list
Evals bootstrap /agents-course:evals-bootstrap 5 · Evals — mine real failures into cases, generate the trace checker
Loop critic agent agents-course:loop-critic 2 · Loop — a checker outside the model: reviews loop results in a clean context

The skills are plain SKILL.md files, so they also work copied into ~/.claude/skills/ or with any harness that reads the Agent Skills format.